A novel machine learning workflow to optimize cooling devices grounded in solid-state physics
Cooling devices grounded in solid-state physics are promising candidates for integrated-chip nanocooling applications. These devices are modeled by coupling the quantum non-equilibirum Green’s function for electrons with the heat equation (NEGF+H), which allows to accurately describe the energetic and thermal properties. We propose a novel machine learning (ML) workflow to accelerate the design optimization process of these cooling devices, alleviating the high computational demands of NEGF+H. This methodology, trained with NEGF+H data, obtains the optimum heterostructure designs that provide the best trade-off between the cooling power of the lattice (CP) and the electron temperature (Te). Using a vast search space of 1.18x10^5 different device configurations, we obtained a set of optimum devices with prediction relative errors lower 4% than for CP and for 1% Te. The ML workflow reduces the computational resources needed, from two days for a single NEGF+H simulation to 10 s to find the optimum designs.
keywords: III–V semiconductors, Machine Learning, Solid state physics, NEGF,
Publication: Article
1733751719427
December 9, 2024
/research/publications/a-novel-machine-learning-workflow-to-optimize-cooling-devices-grounded-in-solid-state-physics
Cooling devices grounded in solid-state physics are promising candidates for integrated-chip nanocooling applications. These devices are modeled by coupling the quantum non-equilibirum Green’s function for electrons with the heat equation (NEGF+H), which allows to accurately describe the energetic and thermal properties. We propose a novel machine learning (ML) workflow to accelerate the design optimization process of these cooling devices, alleviating the high computational demands of NEGF+H. This methodology, trained with NEGF+H data, obtains the optimum heterostructure designs that provide the best trade-off between the cooling power of the lattice (CP) and the electron temperature (Te). Using a vast search space of 1.18x10^5 different device configurations, we obtained a set of optimum devices with prediction relative errors lower 4% than for CP and for 1% Te. The ML workflow reduces the computational resources needed, from two days for a single NEGF+H simulation to 10 s to find the optimum designs. - Julian G. Fernandez, Guéric Etesse, Natalia Seoane, Enrique Comesaña, Kazuhiko Hirakawa, Antonio Garcia-Loureiro & Marc Bescond - 10.1038/s41598-024-80212-9
publications_en